Carbohydrate-appended tumor targeting iron(III) complexes showing photocytotoxicity in red light

Article abstract of DOI:10.1021/ic4028173

Glucose-appended photocytotoxic iron(III) complexes of a tridentate Schiff base phenolate ligand [Fe(bpyag)(L)](NO₃) (1-3), where bpyag is N,N-bis(2-pyridylmethyl)-2-aminoethyl-β-d-glucopyranoside and H ₂L is 3-(2-hydroxyphenylimino)-1-phenylbutan-1-one (H ₂phap) in 1, 3-(2-hydroxyphenylimino)-9-anthrylbutan-1-one (H ₂anap) in 2, and 3-(2-hydroxyphenylimino)-1-pyrenylbutan-1-one (H₂pyap) in 3, were synthesized and characterized. The complex [Fe(dpma)(anap)](NO₃) (4), having bis-(2-pyridylmethyl)benzylamine (dpma), in which the glucose moiety of bpyag is substituted by a phenyl group, was used as a control, and the complex [Fe(dpma)(anap)](PF₆) (4a) was structurally characterized by X-ray crystallography. The structure shows a FeN₄O₂ core in a distorted octahedral geometry. The high-spin iron(III) complexes with magnetic moment value of ～5.9 μ_B showed a low-energy phenolate-to-Fe(III) charge-transfer (CT) absorption band as a shoulder near 500 nm with a tail extending to 700 nm and an irreversible Fe(III)-Fe(II) redox couple near -0.6 V versus saturated calomel electrode. The complexes are avid binders to calf thymus DNA and showed photocleavage of supercoiled pUC19 DNA in red (647 nm) and green (532 nm) light. Complexes 2 and 3 displayed significant photocytotoxicity in red light, with an IC₅₀ value of ～20 μM in HeLa and HaCaT cells, and no significant toxicity in dark. The cell death is via an apoptotic pathway, by generation of reactive oxygen species. Preferential internalization of the carbohydrate-appended complexes 2 and 3 was evidenced in HeLa cells as compared to the control complex 4. A 5-fold increase in the cellular uptake was observed for the active complexes in HeLa cells. The photophysical properties of the complexes are rationalized from the density functional theory calculations.

Full text of DOI:10.1021/ic4028173

Article
pubs.acs.org/IC
Carbohydrate-Appended Tumor Targeting Iron(III) Complexes
Showing Photocytotoxicity in Red Light
Uttara Basu,† Imran Khan,‡ Akhtar Hussain,† Bappaditya Gole,† Paturu Kondaiah,*,‡
and Akhil R. Chakravarty*,†
†Department of Inorganic and Physical Chemistry, Indian Institute of Science, Bangalore 560012, Karnataka, India
‡Department of Molecular Reproduction, Development and Genetics, Indian Institute of Science, Bangalore 560012, Karnataka, India
S
* Supporting Information
ABSTRACT: Glucose-appended photocytotoxic iron(III)
complexes of a tridentate Schiff base phenolate ligand
[Fe(bpyag)(L)](NO3) (1−3), where bpyag is N,N-bis(2-
pyridylmethyl)-2-aminoethyl-β-D-glucopyranoside and H2L is
3-(2-hydroxyphenylimino)-1-phenylbutan-1-one (H2phap) in
1, 3-(2-hydroxyphenylimino)-9-anthrylbutan-1-one (H2anap)
in 2, and 3-(2-hydroxyphenylimino)-1-pyrenylbutan-1-one
(H2pyap) in 3, were synthesized and characterized. The
complex [Fe(dpma)(anap)](NO3) (4), having bis-(2-
pyridylmethyl)benzylamine (dpma), in which the glucose
moiety of bpyag is substituted by a phenyl group, was used as a
control, and the complex [Fe(dpma)(anap)](PF6) (4a) was structurally characterized by X-ray crystallography. The structure
shows a FeN4O2 core in a distorted octahedral geometry. The high-spin iron(III) complexes with magnetic moment value of
∼5.9 μB showed a low-energy phenolate-to-Fe(III) charge-transfer (CT) absorption band as a shoulder near 500 nm with a tail
extending to 700 nm and an irreversible Fe(III)−Fe(II) redox couple near −0.6 V versus saturated calomel electrode. The
complexes are avid binders to calf thymus DNA and showed photocleavage of supercoiled pUC19 DNA in red (647 nm) and
green (532 nm) light. Complexes 2 and 3 displayed significant photocytotoxicity in red light, with an IC50 value of ∼20 μM in
HeLa and HaCaT cells, and no significant toxicity in dark. The cell death is via an apoptotic pathway, by generation of reactive
oxygen species. Preferential internalization of the carbohydrate-appended complexes 2 and 3 was evidenced in HeLa cells as
compared to the control complex 4. A 5-fold increase in the cellular uptake was observed for the active complexes in HeLa cells.
The photophysical properties of the complexes are rationalized from the density functional theory calculations.
The higher rate of glucose metabolism is a general
characteristic of tumor cells.6,7 The overexpression of glucose
transporter GLUT-1 is well documented for a large variety of
tumors and tagging a sugar moiety to the ligand could ensure
higher uptake of the metal complex in tumor cells while
increasing its aqueous solubility.6−9 Glycoconjugation is
reported to be an effective way of increasing the interactions
of the conjugates with lectin type receptors that are
overexpressed in certain malignant cells.10−14 Sugar-dependent
photocytotoxicity and cellular uptake using glycoconjugated
porphyrins and chlorins are reported.15,16 Recently, function-
alization with mannose of photosensitizable nanoparticles is
shown to give enhanced PDT effect due to an active
endocytosis mediated by a membrane receptor.17,18
INTRODUCTION
■
Photodynamic therapy (PDT) has emerged as a noninvasive
treatment modality of cancer.1,2 The therapeutic efficacy of
conventional PDT is based on the generation of singlet oxygen
as the reactive oxygen species (ROS) on light activation of the
photosensitizer (PS). The PS in its triplet excited state transfers
its energy to molecular oxygen to form singlet oxygen (1O2) via
type-II process. Alternatively, it might undergo electron transfer
reactions to form hydroxyl (HO•) and/or superoxide (O2
)
•−
radicals in a photoredox pathway.3 The cytotoxic ROS causes
oxidative cell damage or death.4 One important aspect in
designing PDT agent is to have selective accumulation of the
PS in the cancerous cells leaving the normal and healthy cells
unaffected. There should be moderately enhanced accumu-
lation of the PS in the tumor tissues and additional level of
selectivity could be achieved by irradiating the tumor cells
leaving unexposed healthy cells unaffected. The targeting
efficacy of a PDT agent could be improved by ensuring its
selective accumulation inside the tumor cells. This can be
achieved by designing complexes having ligands conjugated to
sugar, peptides, aptamers, etc. binding to specific receptors that
overexpress on the cancer cells.5
The conventional PDT agents based on macrocyclic organic
dyes show prolonged skin sensitivity, hepatotoxicity and the
necessity to have a high quantum yield of generation of singlet
oxygen.2 Metal complexes with their tunable coordination
environment and varied spectral and redox properties are
Received: November 11, 2013
Published: January 27, 2014
© 2014 American Chemical Society
2152
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
potent alternatives to these organic dyes.19−28 Besides, the
metal complexes can generate different types of ROS using the
redox property of the transition metal. We have earlier shown
that suitably designed 3d metal complexes are potent
photocytotoxic agents in PDT.29−34 An iron(III) complex is
shown to be PDT active in red light.34 In our continued efforts
to develop new iron(III)-based PDT agents, we have now
designed and studied the photocytotoxic effect of carbohydrate-
appended iron(III) complexes in two different cell lines using
red light. Herein, we present the iron(III) complexes of a
tridentate Schiff base phenolate ligand [Fe(bpyag)(L)](NO3)
(1−3), where bpyag is N,N-bis(2-pyridylmethyl)-2-aminoethyl-
β-D-glucopyranoside (bpyag) and H2L is 3-(2-hydroxyphenyli-
mino)-1-phenylbutan-1-one (H2phap) in 1, 3-(2-hydroxyphe-
nylimino)-9-anthrylbutan-1-one (H2anap) in 2 and 3-(2-
hydroxyphenylimino)-1-pyrenylbutan-1-one (H2pyap) in 3
(Chart 1). [Fe(dpma)(anap)](NO3) (4) having bis-(2-
are studied. The sugar appended complexes show significantly
higher cellular uptake than the control species.
EXPERIMENTAL SECTION
■
Materials and Methods. All reagents and chemicals were
purchased from commercial sources (s.d. Fine Chemicals, India;
Sigma-Aldrich, U.S.A.) and used as such without further purifications.
Solvents were purified by standard procedures.35 Supercoiled (SC)
pUC19 DNA (cesium chloride purified) was purchased from
Bangalore Genie. Tris-(hydroxymethyl)aminomethane-HCl (Tris-
HCl) buffer solution was prepared using deionized and sonicated
triple distilled water. Calf thymus DNA (ct-DNA), catalase, superoxide
dismutase (SOD), 2,2,6,6-tetramethyl-4-piperidone (TEMP), 1,4-
diazabicyclo[2.2.2]octane (DABCO), 2′,7′-dichlorofluoresceindiace-
tate (DCFDA), Hoechst 33258, ethidium bromide (EB), 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), propi-
dium iodide (PI), acridine orange (AO), Dulbecco’s modified eagle
medium (DMEM), Dulbecco’s phosphate buffered saline (DPBS), and
fetal bovine serum (FBS) were purchased from Sigma. The
carbohydrate-appended dipicolylamine ligand was synthesized by a
reported procedure.36 Ligand 3-(2-hydroxyphenylimino)-1-phenyl-
butan-1-one (H2phap) was synthesized by refluxing a mixture of
benzoylacetone and 2-aminophenol in methanol for 3 h.37
Chart 1. Schematic Drawing of the Complexes 1−4
The elemental analysis was done using a Thermo Finnigan Flash EA
1112 CHNS analyzer. The infrared and electronic spectra were
recorded using Bruker Alpha and PerkinElmer Spectrum 650
spectrophotometers, respectively. Emission spectral measurements
were carried out with a PerkinElmer LS 55 spectrophotometer. Molar
conductivity measurements were done using a Control Dynamics
(India) conductivity meter. Cyclic voltammetric measurements were
made at 25 °C on a EG&G PAR Model 253 VersaStat potentiostat/
galvanostat with an electrochemical analysis software 270 using a three
electrode setup comprising a glassy carbon working, a platinum wire
auxiliary and a saturated calomel reference (SCE) electrode.
Tetrabutylammonium perchlorate (TBAP, 0.1 M) was used as a
supporting electrolyte. The electrochemical data were uncorrected for
junction potentials. Electrospray ionization mass spectral measure-
ments were done using Esquire 3000 plus ESI (Bruker Daltonics) and
1
Q-TOF Mass spectrometers. H NMR spectral measurements were
made using a Bruker 400 MHz NMR spectrometer. Flow cytometry
(FACS) measurements were done using Calibur (Becton Dickinson
(BD) cell analyzer) at FL1 channel. Magnetic susceptibility measure-
ment of the complex at 300 K was done with solid sample using
MPMS SQUID VSM (Quantum Design, USA). Inductively coupled
plasma optical emission spectral measurements were carried out using
Perkin-Elmer ICP-OES (Model Optima 2000 DV). Viscometric
measurements were performed with a Schott AVS 310 automated
viscometer. Imaging studies were performed with a Zeiss LSM 510
apochromat confocal laser scanning microscope (AO/EB dual
staining) or Olympus IX 81 fluorescence microscope (PI staining).
Synthesis. The ligands 3-(2-hydroxyphenylimino)-9-anthrylbutan-
1-one (H2anap) and 3-(2-hydroxyphenylimino)-1-pyrenylbutan-1-one
(H2pyap) were synthesized by a three-step general procedure in which
anthracene or pyrene was first acetylated with acetyl chloride. The
acetylated product was refluxed with ethyl acetate in presence of
sodium sand to obtain 9-anthroylacetone or pyrenoylacetone by
Claisen condensation reaction. Finally, 9-anthroylacetone or pyrenoy-
lacetone was refluxed with 2-aminophenol in methanol for 8−10 h to
obtain yellow crystals of the product from the filtrate upon standing at
room temperature.
pyridylmethyl)benzylamine (dpma) in which the glucose
moiety of bpyag is substituted by a phenyl group was used as
a control and [Fe(dpma)(anap)](PF6) (4a) was structurally
characterized by X-ray crystallography. The coordination
environment of the metal is modified from the previously
reported iron(III)-catechol system by blocking the sixth
coordination site of iron(III). This modification ensures better
stability of the complexes under physiological conditions by
protecting it from possible attack by transferrin at the labile
coordination site. Transferrin which has an innate affinity for
iron(III) is expected to leach the metal out from the complex
by attacking through any vacant or labile site. The use of two
different tridentate ligands in the present ternary structure is
expected to prevent leaching of iron from the complex. The
anthracenyl or pyrenyl pendant group as planar aromatic
moiety is expected to increase the DNA binding strengths of
the complexes. Complex 1 is used as a control. Similarly,
complex 4 having the carbohydrate group replaced with a
phenyl group is used to ascertain the role of the sugar moiety in
the cellular uptake. The photocytotoxic properties of the
complexes in visible light (400−700 nm) and red light (600−
720 nm), and their differential uptake properties in HeLa cells
3-(2-Hydroxyphenylimino)-9-anthrylbutan-1-one (H2anap).38
A
solution of anthracene (5 g, 28 mmol) and acetyl chloride (12 mL,
168.3 mmol) in dry CH2Cl2 (100 mL) was cooled in an ice-salt bath.
AlCl3 (7.48 g, 56.1 mmol) was added with vigorous stirring
maintaining the temperature below 0 °C. The solution was stirred
for 1 h at the bath temperature and then gradually warmed to 10 °C.
The brick red solution was poured into a mixture of ice and
concentrated HCl. The product was extracted with CH2Cl2, washed
with water (3 × 50 mL) and dried over anhydrous sodium sulfate.
2153
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
Removal of the solvent gave shiny red oil that solidified upon standing.
The crude product was dissolved in hot ethanol (∼50 mL) and heated
to reflux for 30 min and cooled rapidly to room temperature. The
solution was filtered and heated again. Slow cooling gave orange-
brown crystals of the acetylated product that was filtered and dried in
vacuum (4.4 g, 73%). A solution of 9-acetylanthracene (11 g) in ethyl
acetate (24 mL), dry ether (70 mL) was quickly added to sodium sand
(2.2 g). After the vigorous reaction subsided, the mixture was refluxed
for 4 h, cooled and poured into water (70 mL). The aqueous layer was
acidified with acetic acid and the crude 9-anthroylacetone was
recrystallized twice from EtOH by filtering over activated charcoal.
9-Anthroylacetone (2.62 g, 10 mmol) was dissolved in methanol (50
mL) by warming. To it was added 2-aminophenol (1.09 g, 10 mmol).
The orange mixture was refluxed for 8 h. A yellow precipitate thus
formed was filtered and air-dried. The second crop of product in the
form of yellow-orange crystals was obtained from the filtrate upon
1030 vs, 760 w, 570 m (vs, very strong; m, medium; w, weak; br,
broad). Ultraviolet−visible (UV−vis) spectroscopy in 5% DMF−Tris-
HCl buffer, pH 7.2 [λmax (nm) (ε/ M−1 cm−1)]: 560 (720), 395
(3240), 330 (3500). μeff (298 K) = 5.83 μB.
[Fe(bpyag)(anap)](NO3) (2). Yield: ∼74% (0.65 g). Anal. Calcd for
C44H44FeN5O11: C, 60.42; H, 5.07; N, 8.01. Found: C, 60.33; H, 5.16;
−
N, 8.12%. ESI-MS in MeCN (m/z): 812.38 [M-NO3 ]+. Molar
conductance in DMF (ΛM): 72 S m2 M−1. FT-IR: 3350 br (O−H),
−
2990 w, 2700 w, 1640 m, 1460 m, 1370 m (NO3 ), 1030 vs, 740 w,
550 m. UV−vis in 5% DMF−Tris-HCl buffer, pH 7.2 [λmax (nm) (ε/
M
−1 cm−1)]: 505sh (1240), 392 (6790), 373 (6430), 355 (5390) (sh,
shoulder). μeff (298 K) = 5.85 μB.
[Fe(bpyag)(pyap)](NO3) (3). Yield: ∼70% (0.63 g). Anal. Calcd for
C46H44FeN5O11: C, 61.48; H, 4.93; N, 7.79. Found: C, 61.53; H, 4.75;
−
N, 7.69%. ESI-MS in MeCN (m/z): 836.25 [M-NO3 ]+. Molar
Conductance in DMF (ΛM): 71 S m2 M−1. FT-IR: 3360 br (O−H),
−
1
2990 w, 2700 w, 2500 w, 1640 m, 1470 m, 1380 m (NO3 ), 1030 vs,
standing for a couple of days (Yield: ∼80%). H NMR (CDCl3, 400
740w, 570 m. UV−vis in 5% DMF−Tris-HCl buffer, pH = 7.2 [λmax
(nm) (ε/ M−1 cm−1)]: 490 sh (1400), 393(12920), 372 (14070), 352
(10770), 334 (7690). μeff (298 K): 5.83 μB.
MHz) δ (ppm): 8.470 (s, 1H), 8.216 (t, J = 9.6 Hz, 2H), 8.022 (t, J =
9.6 Hz, 2H), 7.484 (m, 4H), 7.121 (m, 1H), 6.926 (m, 1H), 6.623 (m,
1H), 5.704 (s, 1H), 2.033 (s, 3H) (s, singlet; t, triplet; m, multiplet; J,
coupling constant).
[Fe(dpma)(anap)](NO3) (4). Yield: ∼83% (0.63 g). Anal. Calcd for
3-(2-Hydroxyphenylimino)-1-pyrenylbutan-1-one (H2pyap).39
AlCl3 (10 g) was suspended in dry CH2Cl2 (100 mL) and cooled
below 0 °C in an ice-salt bath. Acetyl chloride (5.15 g) was gradually
added and stirred for 15 min. Pyrene (12.6 g) was added in portions,
and the dark-red reaction mixture was stirred at room temperature for
18 h under N2 atmosphere. Ice cold concentrated HCl (∼5 mL) was
added. The reaction mixture was stirred for 15 min and poured over
ice. The product was extracted with CH2Cl2, washed with water (3 ×
50 mL), and dried over anhydrous sulfate. The solvent was evaporated
to afford a yellow oil, which solidified upon standing. The product was
recrystallized from methanol to get bright yellow crystals (Yield:
C43H36FeN5O5: C, 68.08; H, 4.78; N, 9.23. Found: C, 68.15; H, 4.71;
−
N, 9.27%. ESI-MS in MeCN (m/z): 696.33 [M-NO3 ]+. Molar
Conductance in DMF (ΛM): 70 S m2 M−1. FT-IR: 3210 w, 3000 w,
−
1600 w, 1450 w, 1360 m (NO3 ), 1030 w, 830 vs, 540 s (s, strong).
UV−vis in 5% DMF−Tris-HCl buffer, pH = 7.2 [λmax (nm) (ε/ M−1
cm−1)]: 434 sh (3890), 389 (5860), 369 (5050), 350 (4010). μeff (298
K): 5.86 μB.
X-ray Crystallographic Procedures. The crystal structure of
complex 4 as its PF6 salt (4a) was obtained by single-crystal X-ray
diffraction technique. Crystals were obtained from the slow
evaporation of a methanol−acetonitrile (1:1 v/v) solution of the
complex. Crystal mounting was done on glass fiber with epoxy cement.
All geometric and intensity data were collected at room temperature
using an automated Bruker SMART APEX CCD diffractometer
equipped with a fine focus 1.75 kW sealed tube Mo Kα X-ray source (λ
= 0.71073 Å) with increasing ω (width of 0.3° per frame) at a scan
speed of 5 and 15 s per frame. Intensity data, collected using ω-2θ scan
mode, were corrected for Lorentz−polarization effects and for
absorption.40 Structure was solved by the combination of Patterson
and Fourier techniques and refined by full-matrix least-squares method
using SHELX system of programs.41 The hydrogen atoms belonging
to the complex were in their calculated positions and refined using a
riding model. All non-hydrogen atoms were refined anisotropically.
The molecular view was obtained by ORTEP.42 Selected crystallo-
graphic data are given in Table S1 (Supporting Information). The
CCDC number is 958281.
Cytotoxicity. The photocytotoxicity of the complexes was studied
using MTT assay, which is based on the ability of mitochondrial
dehydrogenases of viable cells to cleave the tetrazolium rings of MTT,
forming dark-purple membrane-impermeable crystals of formazan that
could be quantified from spectral measurement in DMSO.43
Approximately 8000 cells of human cervical cancer (HeLa) cells or
keratinocyte (HaCaT) cells were plated separately in a 96 wells culture
plate in Dulbecco’s Modified Eagle Medium (DMEM) containing 10%
FBS. After 24 h of incubation at 37 °C in 5% CO2 atm, various
concentrations of the complexes dissolved in 1% DMSO were added
to the cells, and incubation was continued for 2 and 4 h in dark. The
media was subsequently replaced with DPBS and irradiated with a
broad band visible light (400−700 nm, 10 J cm−2), using a Luzchem
Photoreactor (Model LZC-1, Ontario, Canada) fitted with Sylvania
make 8 fluorescent white tubes or a red light of 600−720 nm (150 J
cm−2), using Waldmann PDT 1200 L. After photoexposure, DPBS was
removed and replaced with DMEM−FBS, and incubation was
continued for a further period of 20 h in dark. After the incubation
period, 5 mg mL−1 of MTT (20 μL) was added to each well, and
incubation continued for an additional 3 h. The culture medium was
finally discarded, and 200 μL of DMSO was added to dissolve the
formazan crystals; the absorbance of the culture medium at 540 nm
was measured using a Molecular Devices Spectra Max M5 plate reader.
1
∼80%). H NMR (CDCl3, 400 MHz): δ 9.031 (d, J = 8.2 Hz, 1H),
8.342 (d, J = 8.0 Hz, 1H), 8.209 (m, 3H), 8.123 (m, 2H), 8.026 (m,
2H), 2.885 (s, 3H) (d, doublet).
A solution was formed of acetylpyrene (3 g) in ethylacetate (8 mL),
and dry ether (25 mL) was quickly added to sodium sand (0.65 g).
After the vigorous reaction subsided, the mixture was refluxed for 24 h.
Water was added slowly, and the aqueous layer was acidified with
acetic acid when a yellow precipitate formed. It was filtered and air-
dried. The crude product was purified by column chromatography,
eluting with 2% CH2Cl2−MeOH. The solvent was evaporated to get
the pure product pyrenoylacetone as a yellow-orange powder (yield:
∼70%). 1H NMR (CDCl3, 400 MHz): δ 8.782 (d, 9.2 Hz, 1H),
8.262−7.994 (m, 8H), 6.207 (s, 1H), 2.276 (s, 3H). Pyrenoylacetone
(2.86 g, 10 mmol) was dissolved in methanol (50 mL) by warming. To
it was added 2-aminophenol (1.09 g, 10 mmol). The orange mixture
was refluxed for 10 h. The yellow precipitate formed was filtered and
air-dried. A second crop of the product was obtained as yellow-orange
1
crystals from the filtrate upon standing for 2 d (yield: ∼80%). H
NMR (CDCl3, 400 MHz): δ 8.781 (d, 9.2 Hz, 1H), 8.266−8.019 (m,
9H), 7.230−6.912 (m, 3H), 5.947 (s, 1H), 2.061 (s, 1H).
Syntheses of the complexes [Fe(bpyag)(L)](NO3) (L = phap, 1;
anap, 2; pyap, 3) and [Fe(dpma)(anap)](NO3) (4). The complexes
were prepared by a general procedure in which Fe(NO3)3·9H2O
(0.404 g, 1 mmol) in methanol (5 mL) was reacted with the
carbohydrate-appended dipicolylamine (0.405 g, 1 mmol, for 1−3) or
phenyl dipicolylamine (0.289 g, 1 mmol, for 4) in methanol (6 mL)
and stirred at room temperature for 30 min and filtered. The methanol
solution of the ligand, namely, H2phap or H2anap or H2pyap (1
mmol), was subsequently added to the filtrate after treating with 2
equiv of triethylamine. The solution was further stirred for another 2 h
at 25 °C and filtered. The filtrate was evaporated to obtain the
complexes as dark-red solid.
[Fe(bpyag)(phap)](NO3) (1). Yield: ∼78% (0.6 g). Anal. Calcd for
C36H40FeN5O11: C, 55.82; H, 5.21; N, 9.04. Found: C, 55.71; H, 5.29;
−
N, 9.24%. ESI-MS in MeCN (m/z): 712.22 [M-NO3 ]+. Molar
Conductance in dimethylformamide (DMF) (ΛM): 69 S m2 M−1.
Fourier transform infrared (FT-IR) spectroscopy: 3320 br (O−H),
−
2930 w, 2600 w, 2490 w, 1590 w, 1440 m, 1370 m (NO3 ), 1290 w,
2154
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
Cytotoxicities of the complexes were measured as the percentage ratio
of the absorbance of the treated cells to the untreated controls. The
IC50 values were determined by nonlinear regression analysis
(GraphPad Prism 5).
then washed twice with 1.0 mL of DPBS by centrifuging at 4000 rpm
at 4 °C for 5 min. The supernatants were gently discarded, and the cell
pellets were suspended in 200 μL of DPBS containing 10 μg mL−1
DNase-free RNase at 37 °C for 12 h. After digestion of cellular RNA, a
20 μL volume of 1.0 mg mL−1 propidium iodide solution was added to
both the mixtures, which were incubated at 25 °C in the dark for
another 20 min. Flow cytometric analysis was performed using a FACS
Calibur [Becton Dickinson (BD) cell analyzer] at FL2 channel (595
nm), and the distribution of cells in various cell-cycle phases was
determined from the histogram generated by Cell Quest Pro software
(BD Biosciences). Data analysis for the percentage of cells in each cell-
cycle phase was performed by using WinMDI version 2.8.
Ethidium Bromide/Acridine Orange Dual Staining Experi-
ment. The changes in chromatin organization in HeLa cells after
photoirradiation were determined microscopically after treatment with
complexes 2−4 by dual staining method using acridine orange (AO)
and ethidium bromide (EB). Briefly, about 2 × 104 cells were allowed
to adhere overnight on a coverslip placed in each well of a 12-well
plate. The cells were treated with the complexes 2−4 (10 μM) for 4 h
in the dark, followed by irradiation with visible light of 400−700 nm
(10 J cm−2). Dark controls were also used. The cells were allowed to
recover for 1 h, washed thrice with DPBS, stained with an AO/EB
mixture (1:1, 10 μM) for 15 min, and observed with a confocal laser
scanning microscope (Zeiss LSM 510 apochromat).
Propidium Iodide Staining. The changes in chromatin
organization after photoirradiation in HeLa cells were also determined
after treatment with complexes 2−4 by propidium iodide staining.
Similar to the former procedure, about 2 × 104 cells were allowed to
adhere overnight on a coverslip placed in each well of a 12-well plate.
The cells were treated with the complexes (10 μM) for 4 h in the dark,
followed by irradiation with visible light of 400−700 nm (10 J cm−2).
Dark controls were also used. The cells were fixed and permeabilized
using chilled methanol and kept for 10 min. They were treated with PI
(1.0 mg mL−1 in DPBS), and images were recorded with a
fluorescence microscope (Olympus IX 81). The apoptotic cells were
identified by the presence of highly condensed nuclei as opposed to
the untreated cells or cells treated with the complexes but unexposed
to light, which showed intact cellular morphology.
DNA Binding Methods. Viscometric titrations were performed
with a Schott AVS 310 Automated Viscometer at 37 °C. The initial
concentration of ct-DNA was 150 μM in NP (nucleotide pair), and the
flow times were measured with an automated timer. Data were
presented as (η/η0)1/3 versus [complex]/[DNA], where η is the
viscosity of DNA in the presence of the complex and η0 is that of ct-
DNA alone. Viscosity values were calculated from the observed flow
time of DNA-containing solutions (t) corrected for that of the buffer
alone (t0), η = (t − t0)/t0.
DNA melting experiments were carried out by monitoring the
absorption intensity of ct-DNA (200 μM) at 260 nm at various
temperatures, both in the absence and presence of the complexes (20
μM) using a Perkin-Elmer 650 spectrophotometer with a temperature
controller. The absorption band at 260 nm of ct-DNA was measured
as a function of increasing the temperature in the presence and
absence of the iron(III) complexes (20 μM) in phosphate buffer
comprising 5 mM Na2HPO4, 5 mM NaH2PO4, 1 mM Na2EDTA, and
5 mM NaCl (pH = 6.8).
The experiments were carried out in Tris-HCl buffer (5 mM, pH =
7.2) using solutions of the complexes 1−4 in DMF. Calf thymus DNA
(ct-DNA, ca. 250 μM) in the buffer medium gave a ratio of UV
absorbance at 260 and 280 nm of ca. 1.9:1, suggesting the DNA was
apparently free from protein. The concentration of DNA was
estimated from its absorption intensity at 260 nm with ε value of
6600 M−1 cm−1.46 Absorption titration experiments were performed
by varying the concentration of ct-DNA while keeping the complex
concentration as constant. Due correction was made for the
absorbance of ct-DNA itself. The spectra were recorded after
equilibration for 5 min. The intrinsic equilibrium binding constant
(Kb) of the complexes to ct-DNA were obtained by the McGhee−von
Hippel (MvH) method using the expression of Bard and co-workers
by monitoring the change of the absorption intensity of the spectral
Quantification of Cellular Iron by ICPOES. For cellular uptake
studies by inductively coupled optical emission spectroscopy
(ICPOES), HeLa or HaCaT cells were grown in 6 well tissue culture
plates in DMEM supplemented with 10% FBS to 80% confluency. The
complexes 1−4 were dissolved in DMSO and added to the cells to give
a final complex concentration of 10 μM (final concentration of DMSO
1% v/v). They were incubated for 2 and 4 h at 37 °C in a CO2
incubator. The culture medium was discarded, and the cells were
washed thrice with ice-cold DPBS. The cells were harvested, taken up
in DPBS, and centrifuged at 3000 rpm for 15 min. After discarding the
buffer, the cell pellets were dissolved in 70% nitric acid at 65 °C for 2
h. The samples were diluted further to a final concentration of 5%
nitric acid with double-distilled water containing 0.1% Triton X100.
The iron content inside the cells was estimated using the spectrometer
previously calibrated with standard ferric nitrate solutions. The protein
content inside the cells was estimated by Bradford assay after lysing
the cells with 1 N NaOH solution. The quantity of iron was expressed
as ng/mg of protein.
DNA Fragmentation Analysis. The analysis was carried out to
ascertain the apoptotic mechanism induced by the complexes 2 and 3
for cell death. Briefly, 0.3 × 106 cells were taken in 6 well plates, grown
for 24 h and later treated with the complexes (10 μM) for 4 h in dark.
Light was exposed to one of the dishes for 1 h, and again the cells were
left to grow for 4 h, along with their dark control. After 4 h, cells were
trypsinized, washed with DPBS, and resuspended in 0.4 mL of lysis
buffer (10 mM Tris-HCl, pH = 8.0, 20 mM ethylenediaminete-
traacetate (EDTA), 0.2% triton-X 100), with an incubation of 20 min
in ice. Lysed cells were centrifuged for 20 min at 13 000 rpm, and their
supernatants, having soluble chromosomal DNAs, including both high
molecular weight DNA and nucleosomal DNA fragments, were
collected. Phenol/chloroform treatment was performed to remove the
protein present. The supernatant was precipitated with 1/10 volume of
3 M sodium acetate (pH = 5.8) and 2 volumes of ethanol at −20 °C
for overnight. DNA pellet was washed with 70% alcohol and
resuspended in TE containing RNase (1X Tris-EDTA with 100 μg
mL−1 RNase). DNA samples were placed into the well of 1.5% agarose
gel. The agarose gels were run at 70 V for approximately 2 h before
they were photographed under UV light.
DCFDA Assay for ROS Generation. The DCFDA was used to
detect the generation of cellular ROS.44 Cell-permeable DCFDA could
be oxidized by cellular ROS to generate a fluorescent 2′,7′-
dichlorofluorescein (DCF) with an emission maxima at 525 nm.45
The percentage population of cells generating ROS was determined by
flow cytometry analysis in which HeLa or HaCaT cells were incubated
with the complexes 2 and 3 (10 μM) for 4 h followed by irradiation
with visible light (400−700 nm) for 30 min in serum-free conditions.
The cells were harvested by trypsinization, and a single-cell suspension
of 1 × 106 cells mL−1 was made. The cells were then treated with 1 μM
DCFDA solution in DMSO in dark for 15−20 min at 25 °C. The
distribution of DCFDA-stained HeLa cells was determined by flow
cytometry in the FL-1 channel.
Fluorescence Activated Cell Sorting (FACS) Analysis. To
investigate the effect of complexes 2−4 on the cell cycle, 3 × 106 HeLa
cells were plated per well of a 6-well tissue culture plate in DMEM
containing 10% FBS. After 24 h of incubation at 37 °C in a CO2
incubator, DMSO solutions of complexes 2−4 (10 μM) were added to
the cells, and incubation was continued in dark for 2 h. The medium
was subsequently replaced with DPBS, and irradiation was done with
visible light (400−700 nm) using a Luzchem Photoreactor. After
irradiation, DPBS was removed and replaced with DMEM containing
10% FBS, and incubation was continued in the dark for a further
period of 12 h. Cells were then trypsinized and collected into 1.5 mL
centrifuge tubes. The cells were washed once with DPBS (pH = 7.4)
and fixed by adding 800 μL of chilled 70% methanol dropwise with
constant and gentle vortexing to prevent any cell aggregation. The cell
suspensions were incubated at −20 °C for 6 h. The fixed cells were
2155
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
Table 1. Selected Physicochemical Data for Complexes 1−4
a
b
c
d
e
f
g
complex
λ (nm) (ε/M−1 cm1)
IR (O−H) (cm−1
)
ΛM (S m2 M−1
)
Epc (V)
μeff
Kb (M−1
)
ΔTm (°C)
1
2
3
4
560 (720)
505 (1240)
490 (1400)
434 (3890)
3320
3350
3360
69
72
71
−0.56
−0.58
−0.59
−0.59
5.83
5.85
5.83
5.86
(6.4 1.1) × 104
(1.0 0.6) × 105
(2.5 0.7) × 105
1.2
3.5
4.1
3.2
70
(9.3 0.9) × 104
a
b
c
d
In 6% DMF/Tris-HCl buffer (pH 7.2), as shoulder. In solid phase. Molar conductivity values in DMF at 25 °C. Cathodic peak potential of the
e
f
Fe(III)/Fe(II) couple vs SCE in MeCN/0.1 M TBAP at a scan rate of 50 mV s−1. Effective magnetic moment of solid samples at 25 °C. Intrinsic
g
calf thymus DNA binding constant. Change in the calf thymus DNA melting temperature in phosphate buffer (pH = 6.8).
bands with increasing concentration of ct-DNA by regression analysis
presence of triethylamine in methanol (Chart 1). The
carbohydrate moiety was appended to the ligands to enhance
the cellular uptake and solubility of the complex. Complexes
1−3 showed better aqueous solubility than the control species
4, which lacks the carbohydrate moiety. The ligands were
chosen to have a phenolic oxygen to enhance the stability of the
iron(III) complex and to have a low-energy ligand-to-metal
charge transfer (LMCT) band from phenolate π orbital to the
dπ* orbital of iron(III). This LMCT band enabled us to study
the red light-induced DNA cleavage activity and cytotoxicity of
the complexes. We have recently reported an iron(III)
catecholate complex showing red light-induced photocytotox-
icity and nuclear uptake in HeLa cells.34 However, the
disadvantage associated with this catecholate complex is the
presence of a metal-bound nitrate anion, which is found to be
labile, resulting in the formation of an aqua adduct in the buffer
medium. The presence of a labile site for iron(III) complex is
undesirable, considering the high binding affinity of transferrin
to iron(III), making the complex ineffective due to leaching of
the metal.50,51 In the present work, we have tried to circumvent
this problem by incorporating two different tridentate ligands in
the ternary 6-coordinate structure, ensuring higher stability of
the complex in the cellular medium. The pendant anthracene or
pyrene unit is used to introduce planarity to the molecules that
may assist in binding to ct-DNA. As mentioned above, the
carbohydrate moiety is attached to enhance the solubility and
cellular uptake of the complexes. The phenyl analogue 1 was
synthesized and used as a control complex.
2
using the equation (εa − εf)/(εb − εf) = (b − (b2 − 2Kb Ct[DNA]t/
s)1/2)/2Kb, b = 1 + KbCt + Kb[DNA]t/2s, where εf is the extinction
coefficient of free metal complex, εa is the extinction coefficient of the
complex at a given DNA concentration, εb is the extinction coefficient
of the absorption peak arising when the complex fully bound to DNA,
Ct is the total complex concentration, [DNA]t is the concentration of
DNA, and s is the MvH equation-fitting parameter.47,48 The nonlinear
least-squares analysis was done using Origin Lab, version 8.
DNA Cleavage Experiments. Photoinduced cleavage of super-
coiled (SC) pUC19 DNA by complexes 1−4 was studied in red light
of 647 nm and in green light of 532 nm wavelength, using a
continuous-wave (CW) argon−krypton mixed gas ion laser (50 mW
laser power, Spectra Physics model Stabilite 2018-RM). The
experiments were performed in a total volume of 20 μL containing
SC DNA (1 μL, 30 μM) and the complex (20 μM) in a 50 mM Tris
HCl buffer (pH = 7.2) containing 50 mM NaCl. The samples were
irradiated, after which they were incubated for a further period of 1 h
at 37 °C in dark, followed by addition of the loading dye (25%
bromophenol blue, 0.25% xylene cyanol, and 30% glycerol (2 μL)),
and finally loaded on a 1% agarose gel containing 1 μg mL−1 of
ethidium bromide (EB). The gels were run at 50 V for 2 h in Tris−
acetate−EDTA buffer. The bands were visualized in UV light and
photographed, and the intensities of the bands were quantified using
the UVITECH Gel Documentation System. Due corrections were
made for the trace amount of nicked circular (NC) form present in the
original SC DNA and the higher affinity of EB for binding to NC and
to linear forms of DNA over the SC form.49 The error in measuring
the band intensities varied from 3 to 5%. Mechanistic studies were
done using different additives (DMSO, 4 μL; KI, 0.5 mM; TEMP, 0.5
mM; DABCO, 0.5 mM; NaN3, 0.5 mM; and catalase, 4 units) that
were added to SC DNA prior to the addition of the complex. After
light exposure, each sample was incubated for 30 min at 37 °C and
analyzed for the photocleaved products using gel electrophoresis.
Theoretical Studies. The geometry of complex 3 was optimized
by density functional theory (DFT) method using B3LYP/LanLD2Z
level as implemented in Gaussian 09 program. The electronic
transitions with their transition probability were obtained using linear
response time dependent density functional theory (TDDFT).
Complexes 1−4 were characterized from the analytical and
physicochemical data (Table 1). The ESI-MS of the complexes
1−4 in methanol showed primarily the complex cationic peak
− +
[M − NO3 ] , suggesting the stability of the complexes under
experimental conditions (Figures S5−S8, Supporting Informa-
tion). The complexes displayed a low-energy phenolic π to
iron(III) dπ* orbital LMCT band that extends up to the PDT
spectral window of 600−800 nm (Figure 1). The bands for 2
and 3 were largely masked by the high-intensity ligand-centered
bands and appeared as an indistinct shoulder in the absorption
spectrum. The bands in the UV region correspond to the π−π*
transitions of the aromatic anthracene and pyrene moieties.
The IR spectra of the complexes 1−3 displayed a characteristic
broad band near 3000 cm−1, corresponding to the O−H
stretching from the carbohydrate moiety, and the band for the
nitrate anion at ∼1370 cm−1 (Figure S9, Supporting
Information). The complexes are 1:1 electrolytic, giving a
molar conductivity value of ∼70 S m−1 M−1 in DMF. The cyclic
voltammetry measurements, performed in acetonitrile with 0.1
M TBAP, showed an irreversible redox response corresponding
to the Fe(III)/Fe(II) couple near −0.60 V versus SCE (Table
1, Figure S10, Supporting Information). The high negative
value indicates the stability of the Fe(III) over the Fe(II) state.
The irreversible redox responses observed near −1.8 and 1.0 V
RESULTS AND DISCUSSION
■
Synthesis and General Aspects. The ligands N,N-bis(2-
pyridylmethyl)-2-aminoethyl-β-D-glucopyranoside (bpyag) and
3-(2-hydroxyphenylimino)-1-phenylbutan-1-one (H2phap)
were synthesized by reported methods.37 Ligands H2anap and
H2pyap were synthesized in good yields by a general three-step
procedure in which anthracene or pyrene was initially
acetylated, followed by a Claisen condensation reaction with
ethyl acetate in presence of sodium sand. Finally, the ligands
were obtained from Schiff base reaction with 2-aminophenol
(Figure S1, Supporting Information). The ligands were
1
characterized from H NMR spectral data (Figures S2−S4,
Supporting Information). Complexes 1−4 were synthesized in
good yields from the reaction of Fe(NO3)3·9H2O with the
tridentate carbohydrate (for 1−3) or phenyl dipicolylamine
(for 4) derivative and the other tridentate ligand in the
2156
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
Figure 2. (a) Photocytotoxicity of the complexes 2−4 after 2 h
incubation followed by irradiation with visible light of 400−700 nm
(10 J cm−2, green), red light of 600−720 nm (150 J cm−2, red), and in
dark (black) in HeLa (bars 1−3) and HaCaT (bars 4−6) cells, with
the complex numbers in parentheses. Bar 1, HeLa and 2; bar 2, HeLa
and 3; bar 3, HeLa and 4; bar 4, HaCaT and 2; bar 5, HaCaT and 3;
bar 6, HaCaT and 4. (b) DNA ladder of complex 3 upon irradiation in
HeLa cells indicating apoptotic cell death (D = dark, L = light, M =
Marker 100 bp).
Figure 1. The UV−vis spectra of complexes 1−4 in 5% DMF−Tris/
HCl buffer (pH = 7.2). The inset shows an ORTEP view of the
cationic complex [Fe(dpma)(anap)](PF6) (4a) with thermal ellipsoids
at 30% probability level and atom labeling for hetero atoms. The
hydrogen atoms are omitted for clarity.
were due to the Schiff base phenolate ligands. The carbohydrate
ligand did not show any redox response within the
experimental potential window (Figure S11, Supporting
Information). The complexes have a high-spin Fe(III) center,
giving a magnetic moment value of ∼5.9 μB at 25 °C.
measure the PDT effect. Complex 1 did not show any
significant photocytotoxicity in visible light. Complexes 2−4
were remarkably photocytotoxic to both the cells. The IC50
values were ∼10 μM for these complexes in visible light of
400−700 nm in HeLa. The IC50 values in red light were near 20
μM. The dark toxicity of all the complexes were low with high
IC50 values (>80 μM). The carbohydrate conjugates did not
exhibit any substantial enhancement in the cytotoxicity
compared to the phenyl dipicolylamine iron(III) analogue 4
(Table S3, Supporting Information). We envisaged that, due to
glucose being present in the media used for the experiments,
the carbohydrate-appended complexes were not preferentially
internalized by the cancer cells. Thus we explored the
photocytotoxicity of the complexes in these cell lines using
glucose-free medium, and the incubation time was varied with 2
and 4 h. The modification in the experimental procedure led us
to the expected results. The carbohydrate-appended complexes
2 and 3 were seen to be significantly more photocytotoxic in
visible light compared to the nonsugar analogue 4 in HeLa cells
(Table 2, Figures S14 and S15, Supporting Information).
Within 2 h of incubation, complexes 2 and 3 displayed 2- to 3-
fold higher photocytotoxicity compared to the control complex
4. The difference in photocytotoxicity in the cells was more
pronounced when incubation time was 2 h than 4 h. Again this
effect was more prominent in HeLa cells compared to
immortalized HaCaT cells. This could be due to the fact that
initially the complexes are ingested via a receptor-mediated
pathway in HeLa cells, resulting in differential uptake as
observed.15 However, with the lapse of time the complexes
were internalized by diffusion making all of them as taken up
indiscriminately. The iron-uptake experiment was performed to
further substantiate our observation.
X-ray Crystal Structure. Complex 4 was crystallized from a
1:1 (v/v) solution of acetonitrile and methanol as a
hexafluorophosphate salt (4a.MeCN) and characterized by X-
ray crystallography. The ORTEP diagram is shown as inset of
Figure 1 (unit cell packing diagram in Figure S12, Supporting
Information). Crystallographic parameters are given in Table
S1, Supporting Information, and selected bond lengths and
bond angles are given in Table S2, Supporting Information.
The crystals belong to the P21/a space group of the monoclinic
system. The molecular structure of the complex has a distorted
octahedral N4O2 coordination around iron, in which the
N,N,N-donor phenyldipicolylamine and the anthracene-
appended Schiff base bind to the metal in a facial manner.
The Schiff base binds to iron through the enolate and
phenolate oxygen atoms and a nitrogen atom, thus showing
the tridentate O,N,O-donor mode of bonding. The Fe(1)−
O(1) and Fe(1)−O(2) bond distances are 1.899(3) and
1.960(3) Å, respectively. The enolate oxygen is involved in
conjugation with the adjacent double bond, and hence the bond
is longer compared to the phenolate bond. The Fe−N distances
range within 2.0−2.2 Å. The anthracene moiety in the pendant
mode does not show any steric encumbrance.
Solubility and Stability. The complexes were soluble in
methanol, ethanol, acetonitrile, dimethylformamide, and
dimethyl sulfoxide, while being insoluble in hydrocarbon
solvents. Complexes 1−3, having the carbohydrate moiety,
were partially soluble in water. The stability of the complexes
was measured from the time-dependent UV−vis spectral scans.
They were found to be stable in aqueous DMF or DMSO, as
seen from the spectra acquired after different time intervals and
suitable for cellular studies in an aqueous medium (Figure S13,
Supporting Information).
Cellular Iron Content. The photocytotoxicity of complexes
2 and 3 being higher compared to that of 4 for an incubation
time of 2 h could be due to differential uptake of the complexes
in the cancer cells. To ascertain the fact, the iron content inside
HeLa or HaCaT cells was measured by the inductively coupled
plasma optical emission spectroscopy (ICPOES) method after
incubating the cells with complexes 1−4 for 2 and 4 h in dark
(Table 3). It was seen that, for 1−3, the order of iron uptake
followed the cytotoxicity trend of 3 > 2 > 1. The iron uptake by
Cytotoxicity. The photocytotoxicity of the complexes 1−4
along with the ligands were examined in human cervical cancer
cell line (HeLa) and human keratinocyte (HaCaT) cells
(Figure 2, Figures S14, S15, Supporting Information). The
complexes were incubated for 4 h and then irradiated with
either visible light of 400−700 nm (10 J cm−2) or red light of
600−720 nm (150 J cm−2). Dark controls were used to
2157
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
Table 2. A Comparison of the IC50 Values of Selected Compounds in HeLa and HaCaT Cells
HeLa
HaCaT
a
b
a
b
compound
IC50 (μM) (visible light) IC50 (μM) (red light) IC50 (μM) (dark) IC50 (μM) (visible light) IC50 (μM) (red light) IC50 (μM) (dark)
1
78.2 1.2
11.0 0.8
8.2 0.6
40.3 1.2
95.0 0.7
16.6 1.1
17.7 1.0
54.2 1.1
4.3 0.2
>100
87.3 0.7
17.3 0.9
14.5 1.3
47.8 1.1
>100
>100
>100
>100
>100
2
85.4 1.1
87.4 1.2
>100
29.5 1.1
26.2 1.0
55.7 1.3
3
4
cd
,
Photofrin
>41
a
b
Cytotoxicity values correspond to 2 h incubation followed by irradiation in visible light of 400−700 nm (10 J cm−2). Cytotoxicity values
c
correspond to 2 h incubation followed by irradiation in red light of 600−720 nm (150 J cm−2). The values are taken from reference 30 and
d
converted to μM using the approximate molecular weight of Photofrin as 600 g M−1. At 633 nm excitation with fluence rate of 5 J cm−2.
dark for 30 min for controls. The substantial shift in the
fluorescence bands observed for cells treated with the
complexes compared to the cells treated with the dye alone
or the dark controls indicated the generation of ROS (Figure
3(a), Figures S17 and S18, Supporting Information).
Table 3. Cellular Uptake of Iron (ng/mg Protein) in HeLa
and HaCaT Cells for Incubation with 10 μM of the
Complexes 1−4 for 2 and 4 h
sample
HeLa
HaCaT
2 h
4 h
2 h
4 h
cells alone
0.1
4.7
5.2
6.1
3.9
0.2
4.4
4.9
5.1
3.6
1
2
3
4
3.9
4.0
4.9
1.1
1.4
2.3
2.5
1.3
the cells treated with 4 for 2 h incubation was about four to five
times lower than its carbohydrate analogue in HeLa cells. The
effect was somewhat less prominent in HaCaT cells. However,
cells incubated with 1−4 for 4 h showed similar iron content.
Thus the difference in photocytotoxicity observed can be
explained on the preferential internalization of the carbohy-
drate-appended complexes in cancer cells in 2 h by receptor-
mediated endocytosis. With the lapse of time the difference in
amount of the complex ingestion decreases in the cells, giving
similar photocytotoxicity of 2−4, possibly due to cellular uptake
by diffusion mechanism.15
DNA Fragmentation Analysis by Agarose Gel Electro-
phoresis. Apoptotic DNA fragmentation indicates pro-
grammed cell death. Apoptosis is due to the activation of
endogenous endonucleases, leading to cleavage of chromatin
DNA into internucleosomal fragments of about 180 base pairs
(bp) and its multiples thereof. Apoptotic DNA fragmentation
could be analyzed using agarose gel electrophoresis, which
appears as a ladder-like pattern at 180 bp intervals. Necrosis on
the other hand is usually characterized by random DNA
fragmentation that forms a “smear” on agarose gels. Complexes
2 and 3 were tested to ascertain the mode of cell death, using
the fragmentation of genomic DNA. DNA ladder was observed
in HeLa and HaCaT cells on treatment with both the
complexes (10 μM) and on irradiation in visible light of
400−700 nm (10 J cm−2) (Figure 2 and Figure S16, Supporting
Information). However, no such DNA ladder was evidenced
when the cells were treated with the complexes in dark,
indicating a photoinduced cell death via an apoptotic pathway.
DCFDA Assay for Measuring Generation of ROS. We
have used DCFDA to detect cellular ROS.46 Cell-permeable
DCFDA on oxidation by cellular ROS generates the fluorescent
product DCF, with an emission band of 529 nm. The
percentage of cell population generating ROS can be
determined by flow cytometry analysis. In our experiments
we recorded fluorescence of cells alone and those treated with
DCFDA alone or DCFDA and complexes 2−4 after irradiation
with visible light of 400−700 nm for 30 min or incubated in
Figure 3. (a) FACS for ROS generation by complex 3 using DCFDA
dye. Generation of ROS is highlighted by the shift in fluorescence
band positions compared to cells alone in HeLa cells treated with
complex 3, under different experimental conditions as shown by the
color codes. Greater shift implies higher amount of DCF and thus
greater ROS generation. (b) Bar diagram showing cell cycle
distribution (sub-G1, G1, S, and G2/M phases) of the HeLa cells
treated with complexes 2 and 4 (10 μM) and irradiated with visible
light of 400−700 nm (10 J cm−2), along with the dark controls.
FACS Analysis. To examine the role of the complexes in
the cell cycle, the DNA content of the cells treated with
complexes 2 and 4 was measured using a fluorescent dye,
namely, propidium iodide (PI) in a flow cytometer (Figure
3(b), Figure S19, Supporting Information). The experiments
were performed under both light and dark conditions. The
complexes in dark did not induce any significant change
(<10%) in the sub-G1 population (which indicates cell death)
of the cells. Complex 4 on irradiation stimulated modest cell
death (∼29%). However, under identical experimental
conditions, complex 2 showed a marked difference in the
population of sub-G1 phase (∼82%). These figures are in
accord with the trend of cellular uptake. Because of preferential
ingestion of complex 2 in HeLa cells, the percentage of cell
death is nearly 2.5 times higher than that of complex 4. The
2158
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
Figure 4. Confocal microscopy images of HeLa cells treated with complexes 2 (a, d), 3 (b, e), and 4 (c, f) on (a−c) keeping in dark or (d−f)
irradiated with visible light (400−700 nm, 10 J cm−2) after staining with EB/AO (1:1, 10 μM) solution. (i) Fluorescence of AO. (ii) Bright field
images of the cells. (iii) Fluorescence of EB. (iv) The merged images of the former panels. Scale bar in panel (a)-iii is 8.5 μm.
cellular data establish the fact that the carbohydrate-appended
complexes are internalized inside the cancer cells at a faster rate,
compared to the control.
DNA melting measurements, and UV−vis absorption. Viscosity
titrations were conducted using DMF solutions of the
complexes and ct-DNA at 37 °C. EB and Hoechst dye were
used as references as typical DNA intercalator and groove
binder, respectively. From the plot of (η/η0)1/3 versus
[complex]/[DNA], it appears that complexes 2−4 bind to
DNA via partial intercalation mode (Figure S21, Supporting
Information). Complex 1 showed primarily groove binding
propensity to ct-DNA. The DNA melting experiments done in
phosphate buffer (pH = 6.8) to examine the thermal stability of
ct-DNA upon binding to the complexes gave a melting
temperature (Tm) increased by a maximum of 4.1 °C for 3,
and the value of ΔTm ranged within 1−4 °C for the other
complexes (Figure S21, Supporting Information). The
complexes having a pyrenyl or anthracenyl moiety upon
binding to DNA stabilize the double helix and hence require
higher temperature for the strands to break. The equilibrium
binding constants (Kb) were measured from UV−vis spectral
titration of the complexes with ct-DNA in 5% DMF−Tris
buffer (pH = 7.2). The decrease in the absorption of the ligand-
centered bands near 380 nm was monitored for the complexes.
The Kb values were ∼105 M−1, giving the order of 3 > 2 > 4 > 1
(Figure S22, Supporting Information). The planarity of the
pyrene and anthracene groups makes the complexes better
binders to ct-DNA than the phenyl analogue.
DNA Cleavage and Mechanistic Aspects. The chemical
nuclease activity of the complexes (20 μM) was studied in the
presence of external oxidizing (H2O2, 0.5 mM) and reducing
(glutathione (GSH), 0.5 mM) agents using supercoiled (SC)
pUC19 DNA (30 μM bp, 0.2 μg) in Tris-HCl/NaCl buffer (50
mM, pH 7.2) in dark. The Fe(III) complexes on treating with
GSH or H2O2 did not show any significant DNA cleavage
activity. They also did not show any significant hydrolytic
cleavage of DNA (Figure S23, Supporting Information). The
undesired hydrolytic cleavage of DNA is thus prevented by
blocking all the coordination sites of Fe(III) by appropriately
designed ligand systems and, in addition, making the iron(III)
Ethidium Bromide/Acridine Orange Dual Staining. An
EB/AO dual staining experiment was performed in HeLa cells
with the complexes 2−4 to further investigate their differential
uptake property and ability to induce cell death. The
experiment was based on the discrimination of live cells from
the dead cells on the basis of membrane integrity. AO, which
can pass through the plasma membrane, stains the DNA of live
cells and fluoresces green. EB on the other hand is excluded
from the cells having intact plasma membrane and stains the
DNA of dead cells, showing orange fluorescence. The cells
incubated with the complexes 2 and 3 for 2 h and irradiated
with visible light (400−700 nm, 10 J cm−2) showed significant
reddish-orange emission characteristic of the apoptotic cells
(Figure 4). The controls, which were incubated in dark, showed
only prominent green emission. Complex 4 behaved rather
differently than its carbohydrate analogues. Cells treated with 4
under identical experimental conditions did not show any
significant orange emission, thus excluding the possibility of cell
death. This further confirms the differential uptake of the
carbohydrate-appended complexes and their noncarbohydrate
analogue.
Propidium Iodide Staining. A PI staining experiment was
performed to study the change in cell morphology in HeLa cells
upon treating with the complexes 2−4 before and after
irradiation with visible light (400−700 nm, 10 J cm−2).
Significant changes in nuclear morphology, such as chromatin
condensation and cell shrinkage, was observed in the cells
treated with complex 2 or 3 as compared to the evenly stained
nuclear contours of untreated HeLa cells (Figure S20,
Supporting Information). Complex 4 did not induce any such
chromatin condensation. The results indicate that complexes 2
and 3 are preferentially taken up in the cells by a receptor-
mediated pathway, which is not the case for 4.
DNA Binding. Interaction of the complexes 1−4 with ct-
DNA was studied by several methods, specifically, viscosity,
2159
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
complexes as chemical-nuclease inactive by stabilizing the
iron(III) redox state.
mechanism is reported for the iron(III) complexes having
phenolate ligands.52
The photoinduced DNA cleavage activity of the complexes
1−4 was studied using SC pUC19 DNA (30 μM, 0.2 μg) in a
medium of Tris-HCl/NaCl (50 mM, pH 7.2) buffer on
irradiation with red light of 647 nm (50 mW) and green light of
532 nm (50 mW) using a CW argon−krypton mixed gas ion
laser. The extent of DNA cleavage is shown in the gel
electrophoresis diagram (Figure 5). Control experiments using
Theoretical Study. DFT calculations were carried out
using b3LYP/LAN2DZ level with Gaussian 03 programs to get
further insights on the excited-state properties. Calculations
were performed on the geometrically optimized structure of 3
(Figure S26, Supporting Information). The dipole-allowed
transitions for the complex are listed in Table S4, Supporting
Information. This structure showed a highly favored transition
band with a peak in the visible region at 437 nm having
oscillator strength of 0.25. These transitions that were assigned
as LMCT bands involving the phenolate to the iron(III) center
are likely to be responsible for the generation of ROS, resulting
in the photocytotoxicity of the complexes. The high oscillator
Figure 5. Gel diagram showing the photocleavage of SC pUC19 DNA
(0.2 μg, 30 μM) by the complexes 1−4 in Tris-HCl buffer (pH 7.2)
containing 10% DMF on photoexposure to green light of 532 nm (t =
1 h, lanes 2−5) and red light of 647 nm (t = 2 h, lanes 6−9): lane 1,
DNA control; lane 2, DNA and 1 (15 μM); lane 3, DNA and 2 (15
μM); lane 4, DNA and 3 (15 μM); lane 5, DNA and 4 (15 μM); lane
6, DNA and 1 (20 μM); lane 7, DNA and 2 (20 μM); lane 8, DNA
and 3 (20 μM); lane 9, DNA and 4 (20 μM).
strength indicates the probability of formation of charge-
•−
separated reactive species that possibly reduce O2 to O2
,
eventually leading to the formation of hydroxyl radical
species.52 Transitions occurring below 400 nm are ligand-
based, involving the pyrene group. There are mainly three
transitions at 388, 376, and 373 nm having oscillator strengths
between 0.02−0.01. From the molecular orbital pictures it is
seen that for complex 3, the HOMO mainly resides on the
electron-rich pyrene group while the LUMO has significant
contributions from the adjacent aminophenol group (Figure 6).
The proximity of the HOMO and LUMO accounts for the
absence of any significant fluorescence in the complex, despite
having a pyrene group.
the metal salt and the phenolate or the carbohydrate ligands in
light and the complexes in dark did not show any significant
DNA cleavage activity (Figure S24, Supporting Information).
Complex 1, with a pendant phenyl group on the phenolate
ligand, showed modest DNA photocleavage activity. Complexes
2−4 were, however, highly active, showing significant DNA
cleavage activity in both green and red light. Photolysis of SC
DNA with a 15 μM solution of 2−4 in green light of 532 nm
showed ∼80% cleavage of SC DNA on photoexposure for 1 h.
Similarly a 20 μM solution of 2 or 4 induced ∼70% cleavage of
SC DNA on exposure to red light for 2 h. Complex 3 showed
almost 80% cleavage of SC DNA to its nicked circular (NC)
form under similar reaction conditions.
Mechanistic investigations were carried out using complexes
2 and 3. The DNA groove binding propensity of the complexes
was studied using the minor groove binder distamycin-A and
the major groove binder methyl green. Distamycin-A or methyl
green showed ∼10% cleavage of SC DNA in red light of 647
nm. Addition of 2 or 3 to distamycin-A-bound SC DNA did not
show any inhibition in the DNA cleavage activity. However,
addition of methyl green to 2 or 3 significantly inhibited the
DNA photocleavage activity thus suggesting DNA binding of
the complexes via DNA major groove. The complexes did not
induce any significant photocleavage of DNA in argon medium,
indicating the possible involvement of the ROS in the DNA
photocleavage reactions. Addition of singlet oxygen quenchers,
specifically, DABCO, TEMP, and sodium azide, showed no
apparent effect on the DNA cleavage activity, thus ruling out
the possibility of a type-II singlet oxygen pathway, which is
known for the PDT activity of organic macrocyclic dyes like
Photofrin (Figure S25, Supporting Information). In contrast,
hydroxyl radical scavengers, specifically, DMSO, KI, and
catalase, showed significant inhibition in the DNA photo-
cleavage activity in both the light sources used. The mechanistic
data indicate the formation of hydroxide and superoxide
radicals as the ROS in a photoredox pathway whereby metal
reduction follows the photoactivation of the LMCT band,
forming charge-separated Fe2+-L•+ (phenolate−ligand radical
•−
cation) as the reactive species that possibly reduces O2 to O2
Figure 6. The HOMO and LUMO of complex 3 as obtained from the
DFT calculations. The energy of the HOMO is −7.29 eV and that of
the LUMO is −4.28 eV.
by the reactive Fe2+ ion and generates hydroxyl radicals via the
reaction 3O2 + 2H+ → HO• + HO− + 2O2. Such a
•−
2160
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
(6) Macheda, M. L.; Rogers, S.; Best, J. D. J. Cell. Physiol. 2005, 202,
654−662.
CONCLUSIONS
■
This report presents new ternary iron(III) complexes showing
significant photocytotoxicity in red light with low dark toxicity.
The carbohydrate-appended complexes were seen to be
internalized preferentially in the HeLa cells with respect to
the control complex, which lacks the carbohydrate moiety. The
complexes having two different tridentate ligands in a six-
coordinate structure are stable in the aqueous medium. The
observed photocytotoxicity in red light is of importance,
considering greater tissue penetration power of low-energy
light. The complexes caused cell death in an apoptotic pathway
by the generation of ROS on irradiation. The dark toxicity is
found to be significantly low due to the structural stability and
the absence of any hydrolytic DNA cleavage and chemical
nuclease activity. The results are of significance toward
developing targeted metal-based PDT agents in red light
using iron as the bioessential metal ion.
(7) MacIver, N. J.; Jacobs, S. R.; Wieman, H. L.; Wofford, J. A.;
Coloff, J. L.; Rathmell, J. C. J. Leukocyte Biol. 2008, 84, 949−957.
(8) Hanif, M.; Meier, S. M.; Kandioller, W.; Bytzek, A.; Hejl, M.;
Hartinger, C. G.; Nazarov, A. A.; Arion, V. B.; Jakupec, M. A.; Dyson,
P. J.; Keppler, B. K. J. Inorg. Biochem. 2011, 105, 224−231.
(9) Hussein, Y. R.; Bandyopadhyay, S.; Semaan, A.; Ahmed, Q.;
Albashiti, B.; Jazaerly, T.; Nahleh, Z.; Ali-Fehmi, R. Clin. Transl. Oncol.
2011, 4, 321−327.
(10) Ballut, S.; Makky, A.; Chauvin, B.; Michel, J. P.; Kasselouri, A.;
Maillard, P.; Rosilio, V. Org. Biomol. Chem. 2012, 10, 4485−4495.
(11) Zheng, X.; Pandey, R. K. Anti-Cancer Agents Med. Chem. 2008, 8,
241−268.
(12) Zheng, X.; Morgan, J.; Pandey, S. K.; Chen, Y. E.; Baumann, T.
H.; Missert, J. R.; Batt, C.; Jackson, J.; Bellnier, D. A.; Henderson, B.
W.; Pandey, R. K. J. Med. Chem. 2009, 52, 4306−4318.
(13) Sakuma, S.; Otake, E.; Torii, K.; Nakamura, M.; Maeda, A.;
Tujii, R.; Akashi, H.; Ohi, H.; Yano, S.; Mohita, A. J. Porphyrins
Phthalocyanines 2013, 17, 331−342.
ASSOCIATED CONTENT
* Supporting Information
(14) Calvaresi, E. C.; Hergenrother, P. J. Chem. Sci. 2013, 4, 2319−
2333.
■
S
(15) Hirohara, S.; Obata, M.; Ogata, S.; Kajiwara, K.; Ohtsuki, C.;
Tanihara, M.; Yano, S. J. Photochem. Photobiol., B 2006, 84, 56−63.
(16) Hirohara, S.; Obata, M.; Ogata, S. I.; Ohtsuki, C.; Higashida, S.;
Ogura, S. I.; Okura, I.; Takenaka, M.; Ono, H.; Sugai, Y.; Mikata, Y.;
Tanihara, M.; Masao, Y.; Yano, S. J. Photochem. Photobiol., B 2005, 78,
7−15.
(17) Zorlu, Y.; Dumoulin, F.; Bouchu, D.; Ahsen, V.; Lafont, D.
Tetrahedron Lett. 2010, 51, 6615−6618.
(18) Hocine, O.; Gary-Bobo, M.; Brevet, D.; Maynadier, M.; Raehm,
Tables of crystallographic data, bond lengths, and bond angles
for complex 4a, IC50 values of selected compounds in HeLa and
HaCaT cells, excited-state transitions in complex 3, Figures
showing structures in synthesis of ligand H2pyap, NMR spectra
of ligands, ESI-MS and IR spectra of complexes, other Figures
and data as described in text. This material is available free of
charge via the Internet at http://pubs.acs.org.
̀
L.; Richeter, S.; Loock, B.; Morere, A.; Maillard, P.; Garcia, M.;
AUTHOR INFORMATION
Corresponding Author
*E-mail: arc@ipc.iisc.ernet.in. (A.R.C.) Phone: +91-80-
22932533. Fax: +91-80-23600683. E-mail: paturu@mrdg.iisc.
ernet.in. (P.K.) Phone: +91-80-22933259.
■
Durand, J. O. Int. J. Pharm. 2012, 423, 509−515.
(19) Shieh, M. J.; Peng, C. L.; Chiang, W. L.; Wang, C. H.; Hsu, C.
Y.; Wang, S. J.; Lai, P. S. Mol. Pharmaceutics 2010, 7, 1244−1253.
(20) Smith, N. A.; Sadler, P. J. Philos. Trans. R. Soc., A 2013, 371,
20120519 DOI: 10.1098/rsta.2012.0519.
Notes
(21) Schatzschneider, U. Eur. J. Inorg. Chem. 2010, 1451−1467.
(22) Chifotides, H. T.; Dunbar, K. R. Acc. Chem. Res. 2005, 38, 146−
156.
The authors declare no competing financial interest.
(23) Higgins, S. L. H.; Tucker, A. J.; Winkel, B. S. J.; Brewer, K. J.
Chem. Commun. 2012, 48, 67−69.
ACKNOWLEDGMENTS
■
We thank the Department of Science and Technology (DST),
Government of India, and the Council of Scientific and
Industrial Research (CSIR), New Delhi, for financial support
(SR/S5/MBD-02/2007; CSIR/01(2559)/12/EMR-II/2012).
A.R.C. thanks the DST for J. C. Bose National Fellowship
and the Alexander von Humboldt Foundation, Germany, for
donation of an electrochemical system. I.K. and A.H. thank
CSIR for research fellowships. We are thankful to Dr. O. Joy, P.
Pai, and A. Kavya for the FACS analysis, to Deepti Bapat for the
confocal and fluorescence microscopy data, and to K. Mitra for
the ESI-MS data. We thank Prof. Sudhakar Rao, Civil
Engineering Department of our institute for the ICPOES
facility.
(24) Fry, N. L.; Mascharak, P. K. Acc. Chem. Res. 2011, 44, 289−298.
(25) Ford, P. C. J. Am. Chem. Soc. 2009, 131, 15963−15964.
(26) Ishida, M.; Lim, J. M.; Lee, B. S.; Tani, F.; Sessler, J. L.; Kim, D.;
Naruta, Y. Chem.Eur. J. 2012, 18, 14329−14341.
́
(27) Ranyuk, E.; Cauchon, N.; Klarskov, K.; Guerin, B.; van Lier, J. E.
J. Med. Chem. 2013, 56, 1520−1534.
(28) Sgambellone, M. A.; David, A.; Garner, R. N.; Dunbar, K. R.;
Turro, C. J. Am. Chem. Soc. 2013, 135, 11274−11282.
(29) Goswami, T. K.; Chakravarthi, B. V. S. K.; Roy, M.; Karande, A.
A.; Chakravarty, A. R. Inorg. Chem. 2011, 50, 8452−8464.
(30) Saha, S.; Majumdar, R.; Hussain, A.; Dighe, R. R.; Chakravarty,
A. R. Philos. Trans. R. Soc., A 2013, 371, 20120190 DOI: 10.1098/
rsta.2012.0190.
(31) Banerjee, S.; Prasad, P.; Hussain, A.; Khan, I.; Kondaiah, P.;
Chakravarty, A. R. Chem. Commun. 2012, 48, 7702−7704.
(32) Sasmal, P. K.; Saha, S.; Majumdar, R.; Dighe, R. R.; Chakravarty,
A. R. Chem. Commun. 2009, 1703−1705.
(33) Basu, U.; Khan, I.; Koley, D.; Saha, S.; Kondaiah, P.;
Chakravarty, A. R. J. Inorg. Biochem. 2012, 116, 77−87.
(34) Basu, U.; Khan, I.; Hussain, A.; Kondaiah, P.; Chakravarty, A. R.
Angew. Chem., Int. Ed. 2012, 51, 2658−2661.
(35) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of
Laboratory Chemicals; Pergamon Press: Oxford, 1980.
(36) Mikata, Y.; Sugai, Y.; Obata, M.; Harada, M.; Yano, S. Inorg.
Chem. 2006, 45, 1543−1551.
(37) Dey, D. K.; Dey, S. P.; Karan, N. K.; Datta, A.; Lycka, A.; Rosair,
G. M. J. Organomet. Chem. 2009, 694, 2434−2441.
REFERENCES
■
(1) Ethirajan, M.; Chen, Y.; Joshi, P.; Pandey, R. K. Chem. Soc. Rev.
2011, 40, 340−362.
(2) Bonnett, R. Chemical Aspects of Photodynamic Therapy; Gordon &
Breach: London, U.K., 2000.
(3) DeRosa, M. C.; Crutchley, R. J. Coord. Chem. Rev. 2002, 233−
234, 351−371.
(4) Clichici, S.; Filip, A.; Daicoviciu, D.; Ion, R. M.; Mocan, T.;
Tatomir, C.; Rogojan, L.; Olteanu, D.; Muresan, A. Acta Physiol. Hung.
2010, 97, 41−51.
(5) Fan, N. C.; Cheng, F. Y.; Ho, J. A.; Yeh, C. S. Angew. Chem., Int.
Ed. 2012, 51, 8806−8810.
2161
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162

Inorganic Chemistry
Article
(38) Atherton, J. C. C.; Jones, S. J. Chem. Soc., Perkin Trans. 1 2002,
2166−2173.
(39) D’Souza, L. J.; Maitra, U. J. Org. Chem. 1996, 61, 9494−9502.
(40) Walker, N.; Stuart, D. Acta Crystallogr., Sect. A 1983, A39, 158−
166.
(41) Sheldrick, G. M. SHELX-97, Programs for Crystal Structure
Solution and Refinement; University of Gottingen: Gottingen, Germany,
̈
̈
1997.
(42) Johnson, C. K. ORTEP; Technical Report ORNL-5138 for Oak
Ridge National Laboratory: Oak Ridge, TN, 1976.
(43) Mosmann, T. J. Immunol. Methods 1983, 65, 55−63.
(44) Keston, A. S.; Brandt, R. Anal. Biochem. 1965, 11, 1−5.
(45) Takanashi, T.; Ogura, Y.; Taguchi, H.; Hashizoe, M.; Honda, Y.
Invest. Ophthalmol. Visual Sci. 1997, 38, 2721−2728.
(46) Reichman, M. E.; Rice, S. A.; Thomas, C. A.; Doty, P. J. Am.
Chem. Soc. 1956, 76, 3047−3054.
(47) McGhee, J. D.; von Hippel, P. H. J. Mol. Biol. 1974, 86, 469−
489.
(48) Carter, M. T.; Rodriguez, M.; Bard, A. J. J. Am. Chem. Soc. 1989,
111, 8901−8911.
(49) Bernadou, J.; Pratviel, G.; Bennis, F.; Girardet, M.; Meunier, B.
Biochemistry 1989, 28, 7268−7275.
(50) Brown, P. J.; Molloy, C. M.; Johnson, P. M. Placenta 1982, 3,
21−27.
(51) Barbeau, K. Photochem. Photobiol. 2006, 82, 1505−1516.
(52) Maurer, D.; Kraft, B. J.; Lato, S. M.; Ellington, A. D.; Zaleski, J.
M. Chem. Commun. 2000, 69−70.
2162
dx.doi.org/10.1021/ic4028173 | Inorg. Chem. 2014, 53, 2152−2162